A method for capturing a high-quality cardiac plethysmography signal automatically and seamlessly using the video cameras embedded in personal electronic devices, includes a program running in the background that periodically takes a picture of the person using the device, runs face detection and/or recognition algorithm, and upon detection of a face, records a video, and then processes the video using algorithms to assess video quality by extracting video quality parameters. When video quality parameters are above predefined thresholds, the recorded video is processed further to generate a plethysmography signal indicative of cardiac activity. The plethysmography signal may then be processed to deduce cardiac activity. The method maintains a pleasurable user experience with the personal electronic devices.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A method for capturing high-quality cardiac plethysmography signals automatically and seamlessly, comprising: providing a personal electronic device containing a video camera embedded therein and a computer program running in the background that periodically takes a picture of the person using the device with automated video capture functions on, runs at least one of a face detection and face recognition algorithm, and upon detection of a face freezes the automated video capture functions upon stabilization of the automated video capture functions, records a video with the automated video capture functions off, and then processes the video using algorithms which assess video quality by extracting video quality parameters; processing the recorded video further when the video quality parameters are above predefined thresholds, to generate a plethysmography signal indicative of cardiac activity; and optionally, processing the plethysmography signal to deduce cardiac activity, wherein the method maintains a pleasurable user experience with the personal electronic device.
2. The method of claim 1 , wherein the automated video capture functions comprise auto-gain, white-balance or auto-focus.
3. The method of claim 1 , wherein the personal electronic device comprises a smartphone, tablet or laptop.
4. The method of claim 1 , wherein the video camera comprises a front facing video camera.
5. The method of claim 1 , wherein the video quality parameters comprise degree of motion of a detected or recognized face, spectral signature from a detected facial region, and face detection consistency.
6. The method of claim 1 , wherein the cardiac activity comprises average heart rate, heart rate variability or arrhythmias.
7. The method of claim 5 , wherein the degree of motion of the detected or recognized face is assessed by calculating the center of a detected facial region across frames in the video, calculating the average central location across the frames, and then quantifying the average deviation of the centers across the frames from the average central location.
8. The method of claim 5 , wherein the spectral signature is assessed by calculating the average Red, Green and Blue pixel values within a detected facial region in each frame to provide a sequence of 3 values across all frames, converting the average values per frame to Hue to provide a single sequence of numbers across frames, performing a spectral estimation over the sequence to provide a spectrum, identifying the frequencies with the highest spectral peak, and quantifying the identified frequencies prominence compared to the other spectral peaks.
9. The method of claim 5 , wherein the face detection consistency is assessed by calculating the ratio between the frames in the video where a face is detected to the total number of frames in the video.
10. The method of claim 1 , wherein the deduced cardiac activity comprises: calculating the average heart rate (HR) and the Heart Rate Variability (HRV) for every video that exceeds the quality parameter threshold; storing the HR in a database using an entry that comprises User ID, Date of Measurement, Time of Measurement, HR measurement, HRV measurement, and Quality Parameters; and displaying the database.
11. The method of claim 1 , wherein the entire method is implemented on the personal electronic device.
12. The method of claim 1 , wherein a portion of the method is implemented on the personal electronic device and the remaining portion of the method is implemented in a cloud server to which data is uploaded from the device.
13. The method of claim 1 , wherein the method is reduced to practice in the form of a downloadable APP or a program embedded in the device prior to deployment.
14. The method of claim 1 , wherein the method is implemented across multiple devices to gather signals from the same person.
15. The method of claim 1 , wherein an APP is installed by a user on a smartphone, a tablet and a laptop and signals extracted from all devices are aggregated at a remote server to form data of that user.
16. A method for capturing high-quality cardiac plethysmography signals automatically and seamlessly, comprising: providing a personal electronic device containing a video camera embedded therein and a computer program running in the background that periodically takes a picture of the person using the device, runs at least one of a face detection and face recognition algorithm, and upon detection of a face, records a video, and then processes the video using algorithms which assess video quality by extracting video quality parameters which comprise degree of motion of a detected or recognized face, spectral signature from a detected facial region, and face detection consistency; processing the recorded video further when the video quality parameters are above predefined thresholds, to generate a plethysmography signal indicative of cardiac activity; and optionally, processing the plethysmography signal to deduce cardiac activity, wherein the method maintains a pleasurable user experience with the personal electronic device.
17. The method of claim 16 , wherein the degree of motion of the detected or recognized face is assessed by calculating the center of a detected facial region across frames in the video, calculating the average central location across the frames, and then quantifying the average deviation of the centers across the frames from the average central location.
18. The method of claim 16 , wherein the spectral signature is assessed by calculating the average Red, Green and Blue pixel values within a detected facial region in each frame to provide a sequence of 3 values across all frames, converting the average values per frame to Hue to provide a single sequence of numbers across frames, performing a spectral estimation over the sequence to provide a spectrum, identifying the frequencies with the highest spectral peak, and quantifying the identified frequencies prominence compared to the other spectral peaks.
19. The method of claim 16 , wherein the face detection consistency is assessed by calculating the ratio between the frames in the video where a face is detected to the total number of frames in the video.
20. A method for capturing high-quality cardiac plethysmography signals automatically and seamlessly, comprising: providing a personal electronic device containing a video camera embedded therein and a computer program running in the background that periodically takes a picture of the person using the device, runs at least one of a face detection and face recognition algorithm, and upon detection of a face, records a video, and then processes the video using algorithms which assess video quality by extracting video quality parameters; processing the recorded video further when the video quality parameters are above predefined thresholds, to generate a plethysmography signal indicative of cardiac activity; and optionally, processing the plethysmography signal to deduce cardiac activity, wherein the deduced cardiac activity comprises: calculating the average heart rate (HR) and the Heart Rate Variability (HRV) for every video that exceeds the quality parameter threshold; storing the HR in a database using an entry that comprises User ID, Date of Measurement, Time of Measurement, HR measurement, HRV measurement, and Quality Parameters; and displaying the database, wherein the method maintains a pleasurable user experience with the personal electronic device.
21. A method for capturing high-quality cardiac plethysmography signals automatically and seamlessly, comprising: providing a personal electronic device containing a video camera embedded therein and a computer program running in the background that periodically takes a picture of the person using the device, runs at least one of a face detection and face recognition algorithm, and upon detection of a face, records a video, and then processes the video using algorithms which assess video quality by extracting video quality parameters; processing the recorded video further when the video quality parameters are above predefined thresholds, to generate a plethysmography signal indicative of cardiac activity; and optionally, processing the plethysmography signal to deduce cardiac activity, wherein the method maintains a pleasurable user experience with the personal electronic device, wherein a portion of the method is implemented on the personal electronic device and the remaining portion of the method is implemented in a cloud server to which data is uploaded from the device.
22. A method for capturing high-quality cardiac plethysmography signals automatically and seamlessly, comprising: providing a personal electronic device containing a video camera embedded therein and a computer program running in the background that periodically takes a picture of the person using the device with automated video capture functions on, runs at least one of a face detection and face recognition algorithm, and upon detection of a face freezes the automated video capture functions upon stabilization of the automated video capture functions, and processes a video with the automated video capture functions off using algorithms which assess video quality by extracting video quality parameters; processing the video further when the video quality parameters are above predefined thresholds, to generate a plethysmography signal indicative of cardiac activity; and optionally, processing the plethysmography signal to deduce cardiac activity, wherein the method maintains a pleasurable user experience with the personal electronic device.
23. The method of claim 22 , wherein the automated video capture functions comprise auto-gain, white-balance or auto-focus.
24. The method of claim 22 , wherein the personal electronic device comprises a smartphone, tablet or laptop.
25. The method of claim 22 , wherein the video camera comprises a front facing video camera.
26. The method of claim 22 , wherein the video quality parameters comprise degree of motion of a detected or recognized face, spectral signature from a detected facial region, and face detection consistency.
27. The method of claim 22 , wherein the cardiac activity comprises average heart rate, heart rate variability or arrhythmias.
28. The method of claim 26 , wherein the degree of motion of the detected or recognized face is assessed by calculating the center of a detected facial region across frames in the video, calculating the average central location across the frames, and then quantifying the average deviation of the centers across the frames from the average central location.
29. The method of claim 26 , wherein the spectral signature is assessed by calculating the average Red, Green and Blue pixel values within a detected facial region in each frame to provide a sequence of 3 values across all frames, converting the average values per frame to Hue to provide a single sequence of numbers across frames, performing a spectral estimation over the sequence to provide a spectrum, identifying the frequencies with the highest spectral peak, and quantifying the identified frequencies prominence compared to the other spectral peaks.
30. The method of claim 26 , wherein the face detection consistency is assessed by calculating the ratio between the frames in the video where a face is detected to the total number of frames in the video.
31. The method of claim 22 , wherein the deduced cardiac activity comprises: calculating the average heart rate (HR) and the Heart Rate Variability (HRV) for every video that exceeds the quality parameter threshold; storing the HR in a database using an entry that comprises User ID, Date of Measurement, Time of Measurement, HR measurement, HRV measurement, and Quality Parameters; and displaying the database.
32. The method of claim 22 , wherein the entire method is implemented on the personal electronic device.
33. The method of claim 22 , wherein a portion of the method is implemented on the personal electronic device and the remaining portion of the method is implemented in a cloud server to which data is uploaded from the device.
34. The method of claim 22 , wherein the method is reduced to practice in the form of a downloadable APP or a program embedded in the device prior to deployment.
35. The method of claim 22 , wherein the method is implemented across multiple devices to gather signals from the same person.
36. The method of claim 22 , wherein an APP is installed by a user on a smartphone, a tablet and a laptop and signals extracted from all devices are aggregated at a remote server to form data of that user.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
February 8, 2019
February 2, 2021
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.